Department of Foreign Affairs and Trade

• Ministers
• International Issues
• Trade Policy
• Country information
• About Australia
• Embassies
• Home

• Travel advice
• Passports
• Visas

• Today is Lebanon Independence Day

Australia-China FTA Conference in Shenzhen

28-29 June 2006

Day 2: Trade Remedies

Safeguards, Anti-Dumping actions and Countervailing Duties

Martin Richardson, School of Economics, College of Business and Economics, Australian National University

Martin Richardson
School of Economics
College of Business and Economics
Australian National University
Canberra,  ACT 0200
Australia
Martin.Richardson@anu.edu.au

Introduction

In the interests of "fair trade", the GATT/WTO multilateral trade agreement contains a number of provisions that swim directly against its general current and allow countries to protect their import interests.   These provisions fall under the general heading of Administered Protection, meaning that a member country simply has to put in place the relevant administrative structure to operate in accordance with WTO guidelines and the imposition of protective duties or quantitative restrictions then becomes an administrative matter.  This is in contrast to the more usual tariff protection that generally requires legislative intervention each time and requires a country to consider its international obligations in each case.  Consequently, administered protection provides a much easier and less costly route to protection for a firm and these measures have very much become the "new new protectionism."

The WTO/GATT provisions regarding so-called Safeguards, Anti-Dumping (AD) and Countervailing Duties (CVD), find no justification on purely economic grounds.  The one rationale provided for AD - the prevention of predatory behaviour - in fact is not articulated in the GATT provisions at all, and nor is it any longer contained in the enabling legislation of any countries of which I am aware.  (It was a feature of the original US AD provisions, interestingly, but the inability of complainant firms to use the legislation led to its removal after a few years, at which point the legislation became much more popular!)  AD has now become simply a means of protection from efficient rivals or a means of preventing domestic consumers from taking advantage of the largesse of foreign firms or governments.  This is shown clearly in the numbers of AD complaints brought worldwide and also in the nature and practice of AD actions.  The role of CVD is even less economically motivated, if possible: as Dixit has noted, the appropriate response to a foreign subsidy is usually a thank-you note, not a retaliatory measure. 

It is clear, however, that the real purpose of both of these measures in the multilateral system is the same as that which is explicitly recognised in the case of actual Article XIX Safeguards (SG) measures - temporary GATT-legal import restrictions - and that is to make reform tolerable to policy-makers: they are sops to domestic interests that enable countries to implement welfare-improving agreements that might otherwise be rejected by sectoral interests.   In acknowledging this, the question then arises as to how they should be handled in a bilateral deal.  Countries making such deals still have similar political pressures to those faced in the multilateral case (although, typically, these will be more focussed) but one might assume that the very fact that two countries are sitting at a bilateral negotiating table means that there is some recognition of a mutual coincidence of interests and that might mitigate the needs for safeguards against the unpredictable actions of the other. 

In this presentation I want to argue that the Australia-China free trade area (FTA) should seize the opportunity to remove these remedies - and I use the term reluctantly - entirely from the bilateral trade relationship. 

This presentation considers the background to the Australia-China economic relationship in terms of these safeguard measures and suggests some possible approaches to the negotiation of FTA terms concerning these measures.

The problem with ‘safeguards'

It is clear that AD law has nothing to do with preventing predation, either de jure (as noted, countries' enabling legislation for AD never mentions the intent of the dumper) or de facto (the U.S. steel industry, for example, can launch AD complaints against over 30 foreign rivals simultaneously: odd monopolists, these!)  A compelling piece of evidence on this comes from Australia and New Zealand's experience in the Closer Economic Relations (CER) FTA: in a fit of enlightenment both countries abolished trans-Tasman AD from mid-1990, noting that predatory behaviour could continue to be dealt with by the relevant countries' competition policies.  Since that epiphany not a single case of trans-Tasman predation has successfully been prosecuted, to the best of my knowledge. 

So proponents of administered protection generally justify it in terms of fairness: creating a "level playing field."  But as The Economist once tartly observed, continuing the sporting metaphor, the only acceptable evidence of a level playing field seems to be that equal numbers of goals are scored at each end: sectoral trade imbalances, rather than confirming the value of trade, are instead evidence of its unfairness.  The interests of consumers are totally ignored in this and it is interesting that they are almost totally ignored in the negotiations of FTAs as well, so that increased import volumes are always perceived as a cost, not a benefit, of a FTA and differences in productivity are seen as problems, rather than as the very basis for mutually beneficial trade.  But whatever the original intention of provisions such as AD , it is simply not clear that it achieves anything like fairness: "the antidumping law, as it currently stands, has nothing to do with maintaining a "level playing field."  Instead, antidumping's primary function is to provide an elaborate excuse for old-fashioned protectionism."  (Lindsey and Ikenson, 2002 p.1.)

This is apparent both from the explosion of AD initiations and findings worldwide - has trade really become so unfair so suddenly? - and from the practices surrounding implementation of AD codes.  The litany of outrageous practices produced by various countries' AD authorities is long and well-known (see Bovard (1991) for an extensive account of US practices) but again one piece of evidence on this is very suggestive.  When the US and Canada first signed the CUSFTA in 1988 they introduced a disputes settlements procedure under which an ad hoc bi-national panel could be convened on a case-by-case basis, which could review some decision of a country's relevant administrative bodies, but only on the basis of the relevant country's own domestic administrative law.  Such a panel had no power to revoke duties but could simply return decisions that it felt were incorrectly made - by the country's own standards, note - to be re-examined by a country's administrative authorities.  Nevertheless, of 14 cases brought by Canadian firms against U.S. determinations before 1995, a reduction of duties followed in 9 cases.   So, despite the deck being so heavily stacked against foreign respondents in administered protection anyway, nevertheless, domestic agencies still frequently abuse the law in reaching their decisions.

Australia and China

Australia is currently ‘celebrating' one hundred years of anti-dumping, having introduced its first AD legislation in 1906 (the third country in the world to do so, after Canada and New Zealand.)  After a century of practice, it is clear that Australian industries have really got the hang of using AD - according to one study (Messerlin 2004) Australia is by far the biggest user of AD amongst developed countries (as measured by AD measures in force by value of imports, 1995-2001.)  On the other hand, China, by the same author's calculations, is the world's biggest target for AD actions (as measured by AD measures in force by value of exports, 1995-2001.)  Interestingly, however, China is largely a victim of AD levied by other developing countries - only 7% of Australia's AD measures over the period 1995-2001 were imposed on China (which is substantially lower than the corresponding figure for the US (for whom about 15% of AD measures were against China) and the EU (over 21%.)) 

As a main target for AD actions, China would seem to have a very clear interest in reducing the operation of AD laws globally (and, indeed, its trade negotiators have stated just this) but there is a tension here with the global trend in AD use by developing countries, a trend from which China is not exempt.  Figure One shows the number of AD measures initiated by China from 1997 to 2005 and the notable features are, first, the increase over time and, second, the big spurt in initiations since China's WTO accession in late 2001. 

Source: constructed from AD-CHN-v2.0.xls: AD dataset compiled by Chad Bown, http://people.brandeis.edu/~cbown/global_ad/ accessed 07/06/06.

Since that accession, China's AD activity has been quite sector-specific and relatively concentrated against a few countries, as Figure Two illustrates.

 

So, in the abstract, it seems that both China and Australia might have a fair bit to ‘lose' if AD were to be sacrificed: historically Australia has been a very heavy user of AD provisions and China has been a common victim of such provisions but China is increasingly becoming a user itself.  But in terms of AD actions between Australia and China, it is notable that China does not appear to have ever brought an AD case while the WTO reports that Australia initiated 21 AD investigations of Chinese products from 1995 to 2005 inclusive.   Bown's dataset and ACS reports indicate some of the Chinese products investigated by Australia, as shown in Table One.

† A(ffirmative), N(egative), W(ithdrawn), T(erminated), MI(ssing data).

Source: constructed from AD-AUS-v1.0.xls: AD dataset compiled by Chad Bown, http://people.brandeis.edu/~cbown/global_ad/ accessed 07/06/06.  Asterisked items compiled from ACS Initiation Reports, http://www.customs.gov.au accessed 06/06/06.

Ignoring terminated and withdrawn cases and ignoring cases in which no dumping (or injury) was found, we have only 16 cases in Table One of affirmative AD findings by Australian authorities against China from 1989 to 2004.  Again, it does not seem that Australia would sacrifice much by foregoing AD actions against China.

This argument misses two potentially important points, however.  First, AD is renowned for its harassment value and it is not clear that it is appropriate to ignore withdrawn and terminated AD cases - these may well serve the purpose of their initiators nevertheless.  The second point is that the mere existence of the AD architecture may well intimidate foreign firms and deter them from pursuing dumping behaviour.  If so then abolishing the AD system would trigger a lot more dumping behaviour.  Now, one might well respond that this is simply a further benefit of any such abolition but it is not clear that sectoral interests would share that optimistic perspective! 

Notwithstanding these arguments however, recent experience does not suggest that AD has been of critical importance in Sino-Australian trade relations and it does seem to me that AD could be successfully put aside in any FTA with little real cost to either partner.  The situation regarding CVDs and SGs is even clearer, in that there has apparently been no use of them at all between China and Australia.   One might retort that the lack of action on CVDs is a consequence of the lack of transparency in the operation of subsidies.  Indeed, this is an explicit complaint of the U.S.' about Chinese commercial practices, with a recent US Trade Representative review of Chinese practice suggesting that China has failed, "to fully implement its WTO subsidy obligations, particularly with respect to (i) prohibited subsidies and (ii) providing notification of its subsidies to the WTO" (USTR, 2006 p.15.)  To the extent that Chinese legal processes are opaque , however, surely this militates even more strongly for sidestepping them entirely by removing these remedies from a bilateral deal.

A Modest Proposal     

First, I would take issue with the common rationale for administered protection as a political sop to sectors that would otherwise oppose a deal that is in a country's overall interests.  Even to the extent that this argument has some validity, vested in the realpolitik of trade negotiations in practice rather than in the textbook, one might still object that this is a very blunt instrument to deal with the problem.  If we wish to ‘buy off' specific sectors, simple targeting principles tell us that the most efficient instrument to do so will be one that addresses those specific sectors only.  But the architecture of AD and CVDs presents a protectionist mechanism that is available to all industries and sectors and, as such, is not targeted at all.  Now, one might retort that it will only be used by the ‘squeaky wheel' sectors so, de facto, it delivers protection to sectors most likely to be harmed by trade liberalisation, but that simply underlines the fact that a far better approach to buying off sensitive sectors is to do so explicitly: this also has the advantage of making the costs of such special treatment more transparent and therefore more subject to political contest. 

I have already suggested that neither China nor Australia has a lot to lose by completely proscribing administered protection in the FTA negotiations, on the basis of historical usage of these tools.  One concern that has been raised with respect to China's trade policy execution (post-WTO accession as well as prior to 2001), particularly by the U.S., is that procedures are frequently not transparent and appear to be biased in favour of domestic complainants.  For example, the aforementioned USTR Top-to-Bottom Review of US-China trade relations, also notes concerns with China's administration of its AD laws, which it suggests is not consistent with its WTO obligations.  I have already suggested that China is certainly not alone in this (and in Australia, for instance, there is no specific legislation covering the use of Article XIX Safeguard measures), but the key point here is that abolition of bilateral Australia-China AD, SG and CVD provisions completely sidesteps this issue.

Nevertheless, certain industries are likely to object to the removal of their protection from more efficient foreign producers.  Messerlin (2004) notes that, globally, there are a few "antidumping-intensive sectors", notably metals, chemicals, machinery and electrical equipment, textiles and clothing and plastics, which jointly account for 75% of AD measures while constituting less than 50% of world trade (Messerlin, 2004 p.111.)  He suggests that the prevalence of AD actions in sub-sectors characterised by fairly standardised products but frequently oligopolistic structures hints at the use of AD actions by large firms as a means of segmenting international markets.  To the extent there is some truth in this globally, it seems to be reflected in the context of Sino-Australian trade.  Reference back to Table One confirms this: while there are a number of complaints from producers of processed foodstuffs, these have generally been unsuccessful and the main successes have come in chemicals, plastics, glass and steel.  This presents an alternative perspective on AD actions and an alternative interpretation of their welfare consequences.  If their purpose is to facilitate the segmentation of markets internationally, then global welfare will typically be enhanced by their removal.  Messerlin writes,

"In sum, the observed sectoral pattern of antidumping reflects the increasing "privatization" of trade policy by firms that have enough initial oligopolistic power to use the "procollusion" bias …embedded in antidumping regulations - a key lesson that should be kept in mind when implementing these regulations, in China as elsewhere." Messerlin, 2004 p.111.

            Another issue that is pertinent to the Australia-China case lies in the details in China's WTO accession agreement.  There are two significant features of special terms negotiated for that accession.  First, if countries treat China as a Non Market Economy (NME) then they are entitled (for 15 years!) to use proxies (usually prices in third countries) for Chinese home prices or costs in AD cases.  As Messerlin (2004 p.123) notes, this makes the existence of dumping easier to prove.  Second, Section 16 of China's Accession protocol (which applies for 12 years) contains a number of provisions that make the imposition of Article XIX Safeguards on China a lot easier than against other WTO members.  For example, the emergency nature of Article XIX remedies is weakened, with no requirement that they be imposed only in ‘unforeseen circumstances', there is no MFN requirement and the evidence on links between import surges and domestic injury (which need only be ‘material' rather than ‘serious') is weakened (Messerlin, 2004 p.127.)  Furthermore, there is an entirely novel "trade deflection" clause that gives third party WTO members the right to enforce their own Article XIX Safeguard measures on China if they fear that another member's measures will deflect Chinese exports to their own market - to do this requires no investigation or prior notification to be provided! (Messerlin, 2004 p.127.) 

Now, these terms cannot be a bargaining chip for Australia, given that recognising China as a Market Economy was a pre-condition to negotiating a FTA and such recognition involves a permanent waiver of Sections 15 and 16 of China's Accession Protocol, the relevant sections that contain these terms.  But it should be recognised that other countries that do not grant Market Economy status to China do have the option of invoking these terms and the longevity of these clauses - 15 and 12 years respectively for Sections 15 and 16 - means that they could be a significant factor in world trade for some time.   To the extent that there is anything to the trade deflection concerns behind Section 16.8 of the Accession protocol, of course, this means that countries in FTAs with China will be even more subject to such deflection of trade.  Now, to stiffen Article XIX-type Safeguards within the FTA in reaction to this prospect would be a retrograde and very mercantilist prospect; nevertheless, it does suggest that further study of the likelihood of such deflected trade flows would be sensible.

And what of any precedent for treating safeguards in this fashion in a bilateral deal - i.e. removing them from the set of trade remedies available in an intra-FTA dispute?  First, let me clarify that I raise the issue of precedent not to suggest that trade negotiators are somehow bound by the principle of stare decisis but rather just to show that it is not completely out of the blue to consider this kind of issue in a bilateral FTA.  Whether it is appropriate in the Australia-China case is to be determined. 

As noted above, Australia and New Zealand were sufficiently farsighted to remove trans-Tasman AD entirely within the CER FTA but that was a decision taken between two countries with harmonised legal practices and competition policies.   One might object that the systems of China and Australia are sufficiently different that such a far-reaching reform is infeasible.  Again, I would respond that historical usage of AD in this relationship is sufficiently low that it would not appear to be much of a sacrifice to lose AD anyway, even if no alternative anti-predation remedies were made available.  What is more, as the problem of predatory behaviour can be handled by courts internally within the country of destination of a product - that is, Australia can prosecute foreign firms acting anti-competitively within Australia under s.46 of the Trade Practices Act - the only real concern that might arise in negotiations in a FTA is that one's partner will use such domestic remedies inappropriately.  To the extent that there is a lack of transparency in a partner's judicial processes, this is a real concern.  Nevertheless, Australian judicial experience demonstrates that predatory behaviour is very unlikely domestically and it is clearly even less likely in a world market where it would require a global monopoly to be effective. 

As far as Article XIX Safeguards are concerned, there is also already strong precedent in other FTAs that Australia has signed for them to be excluded on a bilateral basis, as in the CER and SAFTA, the Australia-Singapore FTA.  This is a little ironic in the sense that if one had to allow an administered protection mechanism for countries to persuade them to sign trade deals, Article XIX Safeguards are more attractive than the alternatives such as AD.  This is so for a number of reasons but particularly because no ad hoc attempts need to be made to measure price differences internationally and ascribe domestic injury to them.  Nevertheless, it is this relatively less harmful instrument that has most easily been negotiated away!  This example, incidentally, illustrates the point made earlier that precedent may be of limited use in guiding the negotiation of a FTA as each FTA is very much sui generis: we do not have SGs in the deals with Singapore or NZ, we do have them in the deals with the US and Thailand.

An even more modest proposal

If the complete abolition of administered protection in the FTA is considered too extreme a step, an intermediate reform is still attractive.  First, the abolition of Article XIX Safeguards seems innocuous.  To this might be added the retention of AD and CVD measures, but augmented by a CUSFTA-style arbitration procedure.  To preserve notions of economic sovereignty, such a body would, as in the CUSFTA, be empowered to review administrative decisions only on the basis of the relevant country's own administrative law and procedures.  This would have the added attraction of requiring both parties to be explicit about the nature of those procedures - as noted above, the opacity of Chinese procedures and questions of their GATT-consistency have been an issue and this would be resolved under this proposal, at least in the context of Australia-China economic relations.

Furthermore, if sector-specific options are essential to enable the negotiation of a comprehensive FTA, then it would behove negotiators to consider sector-specific safeguards (carefully modelled on the principles of Article XIX to ensure their temporary nature); these at least have the benefit of being targeted, in light of the earlier discussion.

Some further issues

I will finish with a brief mention of two related issues.  The first concerns third parties; in particular, are there concerns that arise in writing a FTA from the fact that the partner countries may each have - now or in the future - separate deals with other countries?  The concern here is one of ‘bilateral opportunism': country A might offer some ‘concession' to country C that induces a FTA but then, in subsequent negotiations with country B, undermines the value of the concession offered to C by offering a similar (or more extensive) concession to B.   Now, in itself this is not a problem - A is liberalising its trade at each step - but the concern arises from the chilling effect that fears of such opportunism might have on C's interest in signing a deal in the first place.  From the Australian perspective, the dominant role of resources in our export mix to China suggests that this "preference diversion" should not be a particularly serious concern and, from both countries' viewpoints, it simply militates in favour of cutting the most favourable deal possible.

The second issue I wish to mention is the interrelationship between international investment flows and trade policy measures, such as the safeguards instruments discussed here.  There is some international evidence that AD actions have triggered AD-jumping international investment (see Belderbos et al (2004) and Blonigen (2002).)  That is, to avoid the imposition of AD duties incurred when exporting to a foreign market, exporters have simply relocated to that market and produced internally.  Again, the pattern of Australian exports to China suggests that this is unlikely from Australian firms and the relative insignificance of the Australian market to China suggests it is unlikely to occur in the other direction.  Nevertheless, any such investment flows are clearly inefficient and again are best avoided by simply removing the barrier which they seek to overcome.

Conclusion

To sum up, the arguments in favour of abolishing AD, CVD and SG actions in the Australia-China bilateral relationship are:

• The ‘principled' argument based on underlying economic welfare that these administered protection devices are standard, welfare-reducing protection; consequently they do nothing for ‘fairness' in that they invite procedural abuse;
• The ‘practical' argument that, to the extent that their current use indicates their value,  they are little used in the bilateral trading relationship so their loss is unlikely to be significant;
• That abolition sidesteps the direct costs of administration of these schemes as well as the mistrust that these procedures can introduce into a relationship (indeed, it has been argued that the central reason for Canada to sign up to the CUSFTA originally was nothing to do with the gains from trade; rather it was the insurance against future fickle US trade policy that the disputes settlement mechanism of the CUSFTA provided);
• That if the rationale for these schemes is political, they are nevertheless very blunt instruments and explicit recognition of ‘sensitive' sectors is a more efficient means of achieving the same political goal;
• That abolition removes the use of AD, in particular, as a ‘facilitating device' to segment international markets.

I have also noted that there are examples of tightening administered protection clauses in FTAs - certainly on the Australian side, at least - and that objections to these arguments should carry little weight, except that more study is needed on the prospective consequences of other WTO members invoking the "trade deflection" terms of Section 16.8 of China's WTO Accession Protocol.  Finally, if the full revocation of these instruments is simply infeasible, I have suggested that a partial reform might also be attractive, involving the general abolition of Article XIX Safeguards but the retention of a very limited number of carefully proscribed sector-specific safeguards, and the retention of AD and CVD actions but with a bilateral oversight panel, along the lines of that established in the initial CUSFTA.

References

Australian Customs Service, 2006, ACS Initiation Reports.  Available at http://www.customs.gov.au.  Accessed 06/06/06

Australian Department of Foreign Affairs and Trade, 2005, Australia-China FTA Joint Feasibility Study.  Canberra: Australia.  Available at http://www.dfat.gov.au/geo/china/fta/feasibility_full.pdf.  Accessed 06/06/06.

Belderbos, R., H. Vandenbussche and R. Veugelers, 2004. Antidumping duties, undertakings, and foreign direct investment in the EU.  European Economic Review 48#2, 429-453.

Blonigen, B., 2002. Tariff-jumping antidumping duties.  Journal of International Economics 57#1, 31-49.

Bovard, J., 1991, The Fair Trade Fraud.  St. Martin's Press: New York.

Bown, C., 2006, Global Antidumping Database Version 2.0.  Mimeo, Brandeis University.

Groombridge, M., 2001, Economic relations after China's accession to the WTO: an American perspective on shared goals and challenges confronting the US and the EU.  Mimeo, Cato Institute, Washington DC.  Available at http://www.cap.lmu.de/transatlantic/download/groombridge.pdf.  Accessed 08/06/06.

Killion, M.U., 2004, Quest for legal safeguards for foreign exporters under China's Anti-Dumping regime.  N.C.J. Intn'l Law & Com. Reg. 29, 417-456. 

Lindsey, B. and D. Ikenson, 2002, Antidumping 101: The devilish details of "unfair trade" laws.  Cato Trade Policy Analysis Paper No. 20, Center for Trade Policy Studies, Cato Institute, Washington DC, USA. 

Messerlin, P., 2004, China in the World Trade Organization: Antidumping and Safeguards.  The World Bank Economic Review 18#1, 105-130.

Prusa, T., 1992. Why are so many anti-dumping petitions withdrawn? Journal of International Economics 33#1, 1-20.

Purchase, K., 2005, 2005 - The Year of the Safeguard? Blake's Bulletin on International Trade - China Focus  December 2005. Available at   http://www.blakes.com/english/publications/InternationalTrade/Dec2005/ InternationalTrade_ChinaFocus.pdf.  Accessed 08/06/06.

Ranjan, P., 2006, Preferential trade areas, multinational enterprises, and welfare.  Canadian Journal of Economics 39#2, 493-515.

Richardson, M., 2006, Third-party anti-dumping: a tentative rationale.  Mimeo, Australian National University.

Trebilcock, M. and R. Howse, 1999, The Regulation of International Trade 2nd Ed.  Routledge: London. 

United States Trade Representative, 2006, U.S.-China trade relations: entering a new phase of greater accountability and enforcement.  Top-to-Bottom Review.  USTR: Washington DC.  Available at http://www.ustr.gov/assets/ Document_Library/Reports_Publications/2006/asset_upload_file921_8938.pdf.  Accessed 06/07/06. 

Vautier, K., and P. Lloyd, 1997, International Trade and Competition Policy: CER, APEC and the WTO. Institute of Policy Studies: Wellington NZ.

World Trade Organisation, 2006, WTO Trade Policy Review, Report by the Secretariat, People's Republic Of China, WT/TPR/S/161 28 February 2006.  Available at http://www.wto.org/english/ tratop_e/tpr_e/tp262_e.htm.  Accessed 09/06/06.

World Trade Organisation, 2006, WTO Statistics on Antidumping.  Available at http://www.wto.org/english/tratop_e/ adp_e/adp_e.htm.  Accessed 07/06/06.